面向视频的人脸特征计算方法  

Video-Based Facial Feature Computation Methods

作　　者：王莹笑 杨彦红 谭云峰 WANG Yingxiao;YANG Yanhong;TAN Yunfeng(School of Information Engineering,Beijing Institute of Graphic Communication,Beijing 102600,China)

机构地区：[1]北京印刷学院信息工程学院,北京102600

出　　处：《应用科学学报》2025年第1期137-153,共17页Journal of Applied Sciences

基　　金：出版学新兴交叉学科平台建设-版权保护与运营管理服务平台;北京市自然基金项目-北京市教委科技计划重点项目(No.KZ202010015021);专业学位研究生联合培养基地建设-电子信息(No.21090224002);北京印刷学院校级项目(No.Eb202404)资助。

摘　　要：本文梳理了近五年视频人脸识别领域的研究成果,对比分析了采用的面向视频的人脸特征计算方法,主要分为传统人脸特征计算方法、深度学习人脸特征计算方法、特征聚合和特征融合方法。传统特征提取方法包括线性的和非线性的,深度学习特征提取方法包括非时空特征提取方法和时空特征提取方法。特征聚合和特征融合方法能够整合多个特征源以及融合不同时间段的特征,提高识别性能。此外,本文还统一分析了相关文献用到的算法、算法的优势、评价指标以及应用,能为实际应用中的视频人脸识别系统提供更可靠和高效的解决方案。This paper presents a review of research in video face recognition conducted over the pastfive years.It provides a comparative analysis of the facial feature computation methods,categorizing them into traditional approaches,deep learning techniques,and feature aggregation/fusion methods.Traditional feature extraction methods include linear and nonlinear approaches,while deep learning methods include spatial and temporal feature extraction techniques.Feature aggregation and fusion methods integrate multiple feature sources and fuse features from different time periods to improve recognition performance.At the end of each subsection,this paper also provides a unified analysis of the algorithms used in the literature,highlighting their advantages,evaluation metrics,and applications.Through this research,we aim to provide more reliable and efficient solutions for practical applications of video face recognition systems and promote further advancements in thisfield.

关 键 词：视频人脸识别 特征提取 特征聚合 特征融合 深度学习 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]